FRAX: A major advance for evaluating osteoporosis, but limitations exist

NAMS Annual Meeting

CHICAGO — The strengths of the WHO fracture risk assessment tool prediction algorithm are many, but limitations in clinical practice remain, two speakers said here at the North American Menopause Society 21st Annual Meeting.

The fracture risk assessment tool (FRAX) is currently the best effort to incorporate clinical risk factors in the determination of fracture risk, according to Bess Dawson-Hughes, MD, senior scientist and director of the Bone Metabolism Laboratory, Jean Mayer USDA Human Nutrition Center on Aging, Tufts University. Advantages of the algorithm include easily determined risk factors, global validation, application in specific regions or nations, and scores that pertain to both men and women.

FRAX is a simple Web tool that evaluates the 10-year probability of a major osteoporotic fracture for both men and women. The algorithm is based on individual patient models and integrates several clinical risk factors (age, weight, family history, prior fracture, smoking, alcohol consumption, corticosteroid use), as well as bone mineral density. In the United States, the cut point for fracture risk above which intervention to minimize fracture risk becomes cost-effective is a 10-year probability of 3% for hip fracture and 20% for other major fractures.

Limitations in calculating risk

FRAX will calculate the 10-year fracture risk score for men and women aged 40 years and older; however, it does not provide recommendations for how to use that information. Absolute risk fracture assessment should provide physicians and patients with a clearer, more accurate description of risk to improve the quality of treatment decisions, Marjorie M. Luckey, MD, medical director, Saint Barnabas Osteoporosis Center, Livingston, N.J., said.

“Although FRAX is a major advance in clinical practice, clinicians should be aware that it can occasionally lead to underestimates or overestimates of fracture risk in some patients,” Luckey said. “For example, it does not accommodate all known risk factors and lacks detail on others.” Known or suspected risk factors that contribute to fracture risk but are not included in the FRAX algorithm include falls, immobilization, epilepsy, chronic obstructive pulmonary disease, diabetes and depression, among others, she said. Other risk factors, like glucocorticoids, alcohol and smoking, have a dose-dependent effect on fracture risk.

Luckey cited another caveat in using FRAX in clinical practice: FRAX score is calibrated only for untreated patients and results can be misleading for patients already taking pharmacologic therapy.

Future improvements

Both speakers said it is essential that physicians are aware of the strengths and limitations of FRAX in clinical practice.

“FRAX will help us make better decisions and it will increase the treatment of very high-risk patients and decrease inappropriate treatment of low-risk patients, but clinical judgment is far from obsolete, and we [as clinicians] still have a job to do in helping our patients make the right decisions,” Luckey told the audience.

An advantage of the FRAX platform, according to Dawson-Hughes, is that it is global and “subject to modification,” which may lead to improvements in the near future. Such additions would include incorporating spine BMD, as well as frequency and history of falls in the algorithm, she said.

“This is a start,” Dawson-Hughes said. “FRAX has a future, and I am optimistic that we will see improvements in this tool as we go forward.” – by Katie Kalvaitis

FRAX is a valuable tool for clinicians that is easy to incorporate into clinical practice and has gained widespread use. In concert with the National Osteoporosis Foundation guidelines, FRAX provides concrete guidance to clinicians and patients alike, both in selecting higher-risk patients for pharmacologic therapy and in reassuring patients at lower fracture risk that pharmacologic treatment is not needed at that time.
However, many limitations of FRAX remain. For example, its predictive value is likely reduced among patients with low lumbar spine bone mineral density but preserved femoral neck BMD. In addition, the FRAX models in their current forms include binary classifications of risk factors that are known to demonstrate dose-dependent effects, and states of rapid bone loss, like early menopuase, are not taken into account. It is clear that FRAX can provide a valuable framework for informing treatment decisions, but clinical judgement must still preside in the end.

– Emily Szmuilowicz, MD

Endocrine Today Editorial Board member

For more information:

• Dawson-Hughes B. New NOF guidelines.
• Luckey MM. FRAX update: issues in practice. Both presented at: North American Menopause Society 21st Annual Meeting; Oct. 6-9, 2010; Chicago.

Follow EndocrineToday.com on Twitter.